2000 SCI. MAR., 64 (Supl. I): 63-71 SCIENTIA MARINA TRENDS IN HYDROZOAN BIOLOGY IV. C.E. MILLS, F. BOERO, A. MIGOTTO and J.M. GlLl (eds.)

Approaches to the ethology of hydroids and medusae (, )*

MARIA PIA MIGLIETTA', LUIGI DELLA TOMMASA1, FRANCESCO DENITTO1, CINZIA GRAVILI1, PATRIZLA PAGLIARA1, JEAN BOUILLON2 and FERDINANDO BOERO

'Dipartimento di Biologia, Stazione di Biologia Marina,Museo dell'Ambiente, Universita di Lecce 1-73100 Lecce, Italy. 2Laboratoirede'Biologie Marine, Universite Libre de BraxeUes, Ave F.D. Roosevelt 50, B-1050 Bmxelles, Belgique.

SUMMARY. The behavioural patems of 26 species of Antho-and Leptome^ invesdgated by vuleorecordi^ The analysed acdvities were. answCTS ^^^ cUgestiSn;Tgesdon7andsmm^mg; The quandty of behavioural p^^^^^ Mt7stodi'ed°so'far'is"sufficient-todemonstTate'that these supposedly "simple" have evolved a complex array of responses to both external and internal stimuli.

Key words: Cnidaria, Hydrozoa, Anthomedusae, Leptomedusae, ethology, feeding.

INTRODUCTION medusae, like Sarsia, the outer nervous ring passes at the base of tentacles, forming a tentacular gan- Cnidarians are the most primitive metazoansglion that receives inputs from the ocelli. This prob- with complex behaviour. The other phyla withably serves as a centre for the directioning of ner- which they are placed at the root of the king-vous impulses between the tentacles and other body dom (i. e., Porifera, Placozoa and Mesozoa) haveregions. Its role, thus, is that of a primitive and mdi- simple reactions to external stimuli only. Cnidarians,mentary brain (Mackie, 1971). Such medusan stmc- furthermore, have a well-developed nervous systemtures have no counterpart in hydroids, whose ner- and some have complex sense organs. They are,vous system is a simple network. Both hydroids and thus, the ideal group to investigate the origins ofmedusae, ani- however, answer a vast array of stimuli. mal behaviour. Hydroid feeding behaviour has not been studied The nervous system of hydromedusae is moremuch, although the first observations date back to complex than the one of hydropolyps, beingthe formed beginning of the 1900. Jennings (1906) was the by a network of cells and fibres forming two ringsfirst to describe as "feeding reaction" a behavioural connected by sparse fibres, one in the exumbrellarpattern consisting in the movement of a hydranth ectodenn and the other in the subumbrella. In some after contact with a prey (tentacle retraction, mouth opening, etc.). Jennings (1906) stated that the feed- *Received July 5, 1999. Accepted January 18, 2000. ing reaction is not a constant reflex, but depends on

HYDROZOAN ETHOLOGY 63 general physiological conditions, and reported on Valuable information on hydrozoan behaviour is the importance of both mechanical and chemicalscattered in literature on other topics. Behavioural stimuli. Many authors described the behaviour ofdata are present in many life cycle studies but are single species, but without attempting a generaloften mentioned neither in the abstract nor in the key classification of behavioural patterns. For instance,words (e. g., Boero et al., 1991). Vannucci (1956) observed that starved Dipurena Most behavioural studies were centred on either hydranths move randomly in all directions whilepolyps or medusae and the present paper is an searching for a prey. Passano (1957) reported onattempt the to identify general behavioural patterns for search movements of the hydranth of Corymorpha.both stages from both Antho- and Leptomedusae, Brinckmann-Voss (1970), followed by Edwards withand or without a medusa stage. Harvey (1983), studied the behaviour of the poly- morphic colony of Thecocodium brieni. Orlov (1997) described the feeding behaviour of SarsiaMATERIALS AND METHODS producta. Marfenin (1981) was the first to recognise the The behaviour of 18 species of Anthomedusae ecological importance of hydroid behaviour. Heand 8 species of Leptomedusae was investigated stated that every species, if examined in sufficient(Table 1). Material was invariably sampled by detail, has a unique feeding behaviour, this beingcollecting hydroids by SCUBA diving. Hydroids, determined, for instance, by the ability in prey cap-fed with Artemia nauplii, were kept into hemi- ture, and by ingestion, digestion, and egestion spherical bowls, either attached to portions of times. Differences in these processes are relatedtheir to natural substrata or transplanted on micro- environmental features, colony architecture, formscopic slides. Hydroids were monitored seasonal- and size of the hydranth and, if present, of the ly in the field to identify fertile periods, and were hydrotheca. collected and brought to the laboratory only Medusae, being able to move in the water col-when medusae were near to be liberated. umn, have more recognisable behavioural patternsMedusae were also kept into hemispherical bowls than hydroids. Numerous authors, thus, studiedand fed with Artemia nauplii. Food was provided medusan swimming and feeding behaviour. A com-daily and water was changed after every feeding mon belief of past researchers (e. g., Hyman, 1940)session. Both hydroids and medusae were kept in was that medusae impact casually with their prey,natural so seawater filtered at 0.45 pm. The bowls that both hydro- and scyphomedusae feed passivelywere kept into thermostatic rooms having both while swimming with their tentacles outstretched.temperature If and photoperiod matching natural this were true, medusan-feeding efficiency wouldconditions. be The behavioural patterns were stud- simply related to tentacle number and length. Manyled under a Leica MZ12 stereomicroscope recent observations on single species, however,equipped led with a videocamera and every pattern to a refined view of medusan behaviours. Mills was recorded on videotape. Reaction to mechan- (1981) and Costello (1988, 1992) distinguishedical stimuli, capture of prey, ingestion, digestion, ambush and cmising predators, relating foragingand egestion were investigated for both hydroids strategies to both tentacle morphology and arrange-and medusae; swimming patterns were studied ment and umbrella shape. Zamponi (1985) for medusae. described an "alimentary space" and measured a "predation cylinder" from both tentacle number and length, and umbrellar diameter: a discrete volumeRESULTS of water in which a medusa can find a prey. Prey qual- ity and quantity, furthermore, can induce varied for-The observed behavioural patterns differed in the aging behaviours; physiological adaptations to lowexamined species, but it was possible to identify food availability are important aspects in hydrome-types of behaviour that are usually present in more dusan ecology and account for the predatory effi-than one species (Table 1). Although polyps and ciency of medusae (Costello, 1992). A particularmedusae constitute a single biological unit, they will case is that of the benthic and crawling medusaebe of considered separately since their behaviours fall Eleutheria, whose behaviour has been describedinto in different categories even within the same t detail by Hadrys et al. (1990). species.

64 M.P. MIGUETTA et al. TABLE 1. - Behavioural patterns ofpolyps and medusae (see text for further explanation and references).

Species Polyp Medusa Mechanical stimuli Prey capture Digestion Swimming Prey capture

Anthomedusae FiUfera Bougainvillia sp. Escape Passive Decrease of feeding space Move Mouth to tentacle Turritopsis nutricula Escape Passive Decrease of feeding space Move Mouth to tentacle Eudendrium spp. Escape Passive Inhibition of capture abilityAbsent Absent Hydractinia sp. Escape Passive Decrease of feeding space Absent Absent Thecocodium brieni Approach Passive Decrease of feeding space Absent Absent Amphinema dinema Unidirectional Passive Decrease of feeding space Move Mouth to tentacle bending Codonorchis octaedrus Escape Passive Decrease of feeding space Move Mouth Octotiara russelli Unidirectional Passive Not seen Not seen Mouth bending Stomotoca atra Unidirectional Passive Not seen Not seen Mouth bending Hydrichthys minis Not seen Not seen Not seen Move Mouth Cladonema radiatum Approach Active Decrease of feeding space Move and catch Tentacle to mouth Coryne producta Approach Active Decrease of feeding space Absent Absent Dipurena halterata Approach Random Decrease of feeding space Not seen Not seen Eleutheria dichotoma Approach Passive Decrease of feeding spaceCrawling Tentacle to mouth Ectopleura wrighti Approach Passive Inhibition of ingestion ability Not seen Mouth to tentacle Cladocoryne floccosa Approach Passive Decrease of feeding space Absent Absent Zanclea sessilis Approach Passive Decrease of feeding space Move Mouth to tentacle Zanclea giancarloi Approach Passive Decrease of feeding space Move Mouth to tentacle Leptomedusae Aequorea forskalea Escape Passive Decrease of feeding space Catch Not seen Mitrocoma annae Escape Passive Decrease of feeding space Not seen Not seen Campalecium medusiferumUnidirectional Passive Decrease of feeding space Move (young) Mouth to tentacle bending Halecium pusillum Escape Passive Not seen Absent Absent Aglaophenia octodonta Escape Passive/Active Inhibition of capture abilityAbsent Absent Plumularia setacea Escape Passive Inhibition of capture ability Absent Absent Clytia linearis Escape Passive Decsease of feeding spaceCatch Mouth to tentacle Obelia dichotoma Escape Passive Inhibition of capture ability Catch Mouth to tentacle

Hydroids with a theca, this behaviour resulted in retraction into it. Well-fed hydranths of Aequorea forskalea Mechanical stimuli and M.itrocoma annae, however, were not able to contract into their thecae and remained outside them Single hydranths were stimulated mechanicallyeven with when contracted. The haleciids, having a forceps, so to simulate harsh contact with a sourcereduced theca, resembled filiferans in this behav- of "disturbance". The behavioural responses fellioural pattern. within three categories: escape, approach, unidirec-3 - tentacle folding. Eudendrium spp. were not tional bending. able to contract neither tentacles nor hydranth; dis- turbed hydranths folded their uncontracted tentacles Escape reactions (Fig. 1). Escape reactions aroundwere the peduncled hypostome. typical of the hydranths of filiferans and most the- cates. They were classifiable in three patterns Approach reactions (Fig. 1). All Capitata, and the according to both morphological constraints anddactylozooids the of the filiferan Thecocodium brieni, t intensity of the stimulus: bent their polyps towards the source of the mechan- 1 - tentacle contraction. Light stimuli usually ical stimulus, whatever its intensity. These species caused tentacle contraction accompanied by tentaclehave short, capitate tentacles that are only slightly folding towards the mouth. Halecium pusillumcontractile. was This active movement was related to the the only species that folded the tentacles towardsway the these species catch their prey (see below). column. 2 - hydranth contraction. Strong stimuli, or reit- Unidirectional bending (Fig. 1). The hydranths erated light stimuli, caused tentacle contractionof and, some pandeids and of the thecate Campalecium . simultaneously, hydranth contraction. In the speciesmedusiferum, when touched, bent in a constant

HYDROZOAN ETHOLOGY 65 Mechanical stimuli lltj^l^ ^ Escape Approach Unidirectional bending Capture U fc^^<^ l-l ^ Passive Active ^ Random ^ Behaviour during digestion^ l.nL n ^ Decrease of feeding space Inhibition of capture

Behaviour during egestion

multiple step \^ expulsion SL v ^ Mouth opt ig peduncled/; Q \w/ hypostomev<.

5 f ff /i ^ ^ f\ n ^ *s' '/ ^ "-^ m Hydranth/manubrium one step . eversion expulsion ^

FIG. 1. Answer to mechanichal stimuli, capture and behavior during digestion in hydroids; behavior during egestion in hydroids and medusae. direction, irrespective of the direction of the stimu-Feeding patterns lus. The hydranth, furthermore, bent at a well- defined place in its column; this allows to hypothe-All feeding patterns were observed while using sise the presence of a muscular "articulation"Artemia func- nauplii as food source. None of the investi- tioning as a knee or elbow. The histology of gatedthis species will ever encounter such prey in the region is not known in detail. wild and it is known that hydroids do not feed on

66 M.P. MIGLIETTA et al. crustaceans only (Barange, 1988; Barange andbeen Gili, a waste of both energy and nematoblasts. Two 1988; Coma et al., 1995; Gili et al., 1996). It is pre-main behavioural patterns were observed during sumed, however, that the feeding behavioural pat- prey digestion: tems remain rather constant whatever the offered food is. The observed feeding patterns fell into threeDecrease of feeding space. Full hydranths with- * categories: out mechanoreceptors tend to contract their body and tentacles, so to minimise the feeding space. This Passive (Fig. 1). This was the typical capturestructural modification was particularly evident in strategy of hydranth with long and usually filiformsmall hydranths like Codonorchis octaedrus which, tentacles extended in the water, defining a wideduring digestion, became so small to nearly disap- feeding space. They wait for the prey to collide pearwith against the substrate (Boero et al; 1997). tentacles after having entered their feeding space.Mechanoreceptors, when present, shmnk to little This feeding strategy was observed in most of thebuttons. This lowered the sensitivity to water move- investigated species. ment and hydranths did not bend toward approach- Thecocodium brieni had a peculiar passive-activemg prey. behaviour. The prey was passively captured by the dactylozooids. The gastrozooids, then, took it Inhibition of capture (Fig. 1) or of ingestion. actively from the dactylozooids (see Brinckmann- Some hydranths did not capture any prey during Voss, 1970 and Edwards and Harvey, 1983 fordigestion, even if their tentacles remained extended detailed descriptions). in the water. This suggests that nematocyst dis- charge is under some kind of control (Bumett et al., Active (Fig. 1). The active strategy occurred typ-1960; dark and Cook, 1986; Grosvenor and Kass- ically in hydranths with mechanoreceptors (TardentSimon, 1987), being prevented when the enteron is and Schmid, 1972); these were able to feel an full. This inertia, however, could also be the answer approaching prey, bent actively towards it and to a temporary lack of nematocysts after many cap- caught it with their tentacles. The hydranths per- tures. forming this behaviour usually had short capitate Well-fed hydranths of Ectopleura wrighti did not tentacles, but their ability to "move" around the ingeststalk prey and did not move. The long aboral tenta- led to the control of a hemispherical feeding space.cles, however, continued to catch prey and kept it Another active feeding pattern was that of theuntil the coelenteron was at least partly empty. Prey hydranth of Aglaophenia octodonta whose relative-items, then, were picked up, one by one, by the oral ly long and filiform tentacles beat the surroundingtentacles. water creating a flow that conveyed little prey or organic matter towards the hypostome. Medusae

Random (Fig. 1). This behaviour was typical of The number of observations for medusae was Dipurena halterata, whose hydranths bent random-lower than for hydroids since many of the investi- ly in all directions exploring a hemispherical space,gated species had no medusae. For some species from which they captured every passing prey. Thewith medusae, furthermore, it was impossible to hydranth of D. halterata can also feel water vibra- observe behavioural patterns due to rearing difficul- dons and bend towards the source. Unfed polypsties. Hydranth growth is rapid and there are no dis- bent continuously to increase the probability to tinctcon- stages, although tentacle number increases in tact a prey, but the swinging frequency decreasedthe ingrowing athecate hydranths, possibly influenc- fed hydranths. In the latter case, hydranths benting only their possibilities of prey capture. Medusae, when excited by a swimming prey. instead, usually change much more than hydranths during growth. The medusae of both Clytia and Sar- Behaviour during digestion sia, for instance, are almost round at release, where- as they become respectively more flat or more elon- The main function of hydranths was to capturegate at later growth stages. The velar opening of prey with their tentacles and nematocysts and tonewly released medusae is usually small and jet ingest it. Once the coelenteron was full, the capturedpropulsion is intense. Larger medusae often have food that would have remained unused would have much wider velar openings and their propulsion is

HYDROZOAN ETHOLOGY 67 possibly due to flapping (Costello and Ford, in was similar to the one reported for the hydranths of press). These changes surely influence behaviouralAglaophenia octqdonta. Boero and Sara (1987) patterns. In the present paper, we report mainlydescribed on the same behaviour for the medusae of fully-grown medusae. Obelia longissima. 2 - Tentacle to mouth. Captured preys are Swimming behaviour brought actively to the mouth by the tentacles. In Cladonema radiatum, prey items were forced into Swimming to move (Fig. 2). This behaviouralthe pat- subumbrellar cavity by the motile and branched tem corresponds to the ambush predation describedmarginal tentacles that passed them to the oral ten- by Mills (1981). Medusae swam only to changetacles the which passed them to the mouth. In Eleuthe- foraging site and prey. They remained still in eitherria dichotoma the casual contact of tentacles with a a horizontal (Codonorchis, Amphinemd) or verticalprey during crawling induced active tentacle move- position (Bougainvillia, Zanclea, Turritopsis andment towards the mouth, allowing prey ingestion. Campalecium), or while resting on the bottom (Cladonemd). With mouth (Fig. 2). Some species captured preys with their lips and not with their tentacles. Stomoto- Swimming to catch (Fig. 2). This behaviour cor-ca atra and Octotiara russelli belong to this catego- responds to that of cruising predators (Mills, 1981).ry (Boero and Bouillon, 1989). Hydrichthis mirus Medusae move actively in the water column search-anchored one or two tentacles to the bottom, then it ing for prey. Capture occurs when the prey contactsswam away while remaining attached, so stretching the tentacles. Capture rates depend on swimmingthe attached tentacles for a rather long distance. velocity, prey type and predator morphological char-Subsequent contraction of the attached tentacles acteristics (Mills, 1981; Costello, 1988; 1992). caused a rapid backward movement of the umbrella, Cladonema medusae can catch prey also whileduring which preys entered the subumbrellar cavity swimming as described by Rees (1979) but alsoand can contacted with the oral lips (Boero et al; 1991). remain anchored to the bottom for a long time, so Codonorchis octaedrus used its lips to capture falling in the preceding category. prey that entered by chance in its subumbrellar cav- ity (Boero et al., 1997). Crawling (Fig. 2). The medusae are anchored to the bottom with adhesive tentacular pads, and preyEgestion in hydroids and medusae (Fig. 1) is captured with the tentacles. The change of forag- ing site occurs by walking on the bottom with the Undigested food remains were eliminated during tentacles. This behaviour was observed by Hadrys etegestion, allowing further uptake of food. Hydroids al. (1990) in Eleutheria dichotoma. and medusae had similar egestion behaviours (per- formed by hydranths and manubria respectively), Prey capture and ingestion and this section covers both stages. The elimination of small amounts of waste occurred by contraction Prey capture occurs with two main patterns: of manubrial/hypostomial walls, resulting in push- ing undig^sted material towards the mouth, which With tentacles (Fig. 2). Two patterns of prey was then opened to allow egestion. The oral tenta- ingestion were recorded: cles, when present, were turned downwards in 1 - Mouth to tentacle. Tentacles captured the hydranths and upwards in medusae. When waste prey and brought it to the subumbrellar margin bywas more abundant, this behaviour was often fol- a fast retraction. After this, the manubrium moved lowed by a turning inside-out of the distal part of the out of the subumbrellar cavity to picked up the hydranth/manubrium, so that the digestive wall was prey. Crumpling was a particular case of mouth-to-at least partly exposed. This behaviour was com- tentacle reactions: the umbrella contracted, pushedmonly observed in medusae, during cmmpling, but the manubrium out the velar opening, and the was not observed in the hydranths of the observed mouth ingested the prey. In Obelia dichotoma species of Turritopsis, Halecium, and Plumularia. preys were killed by mouth nematocysts but were The hydranths of the Campanulariidae had par- the continuous tentacle movement that brought ticular egestion patterns, linked to the presence of a them towards the mouth. This active filter feedingseparation between the hypostomic and the gastric

68 M.P. MIGLIETTA et al. Capture behaviour

Mouth to tentacle

Cmmpling Tentacle to mouth ^ Capture with tentacles

y y ^

Capture with mouth

Swimming to catch

^ c / ^ Swimming to move

^ Swimming behaviour

FIG. 2. - Capture and feeding behavior in medusae. cavity, resulting in the so-called peduncled or trum-tome while the sphincter was open. The sphincter pet-shaped hypostome. This constriction acted contracted,as a causing fragmentation of the egesta, part sphincter that played a paramount role in the func-of the egesta remained in the hydranth coelenteron, tioning of the hydranth cavity. Three main wayspart of in the hypostome. The mouth was opened and functioning of peduncled hypostomes were recog-the constriction of the hypostome walls caused eges- nised: tion. The expulsion of the rest of the egesta occurred 1 - Multiple steps expulsion of egesta. Part ofin the further steps, identical to the first one. This behav- undigested material was passed towards the hypos-iour was observed Clytia linearis.

HYDROZOAN ETHOLOGY 69 2 One step expulsion of egesta. The mouth waswas observed in both athecates (many Pandeidae opened while the sphincter was closed, allowingspecies) the and thecates (Campalecium medusiferum). entrance of water into the hypostome. The mouthThe presence of the same behaviour in such differ- / was closed and the sphincter was opened, the hypos-ent groups suggests convergent evolution that led to tome walls were contracted, pushing water into athe still uninvestigated flexible zone. hydranth coelenteron. The incoming water pushedThe behaviour of medusae ranges from "classi- up the egesta while the hydranth walls contracted;cal" patterns of prey capture, to others such as the the egesta were passed into the hypostome. Theactive filter feeding of Obelia (described by Boero sphincter contracted, separating the hypostomeand from Sara, 1987) or the yo-yo tentacle contraction of the hydranth coelenteron; the mouth was opened,Hydrichthys (described by Boero et al., 1991). A allowing waste elimination. This behaviour wasworld apart is the behaviour of symbiotic hydroids observed in Obelia. such as Halocoryne (described by Piraino et al., 3 - Hypostome eversion. In Eudendrium waste1991) and Eugymnanthea inquilina (described by was pushed up in the hypostome through the sphinc-Piraino et al., 1994), reaching extremely high levels ter, but then it was egested by the turning insideof out complexity and specialisation. of the distal part of the hydranth. Scattered unpublished observation, furthermore, suggest that the behavioural patterns of both hydroids and medusae are even more diverse than DISCUSSION presented here and that the available observations cover just a minor part ofhydrozoan ethology. Both The feeding patterns of capitate hydroids ethologydepend and behavioural ecology focused much on on tentacle number and on presence/absence of"higher" invertebrates and vertebrates but the avail- mechanoreceptors. Passive predators have a highable evidence suggests that even "lower" groups can number of tentacles controlling a wide feeding space perform behaviours of high complexity. in order not to fail the capture of a prey passing in their vicinity (e. g., Zancled). Active predators (e. g., Cladonema, Sarsia, Coryne) have mechanorecep-ACKNOWLEDGEMENTS tors sensitive to prey movement. Their feeding space is wide due to the possibility of hydranth bendingWork done with contributions ofM.U.R.S.T. (40 towards approaching prey, so that even a low num-and 60% programmes), C.N.R. (PRISMA 2 Pro- ber oftentacles are sufficient to kill an alreadyject), locat- the Amministrazione Provinciale di Lecce, ed food item. N.S.F. (P.E.E.T.) and of the F.N.R.S. de Belgique. The feeding behaviour of Dipurena halterata is intermediate between passive (without mechanore- ceptors and many tentacles) and active (with REFERENCES mechanoreceptors and few tentacles) patterns, with random bending movements, increasing theBarange, feeding M. - 1988. Prey selection and capture strategies of the bentluc hydroid Eudehdrium racemoswn. Mar. Ecol.Prog. Ser. space. Dipurena halterata has fewer tentacles than 47: 83-88. Zanclea (passive predator) but more than SarsiaBarange, or M. and J.M. Gili. - 1988. Feeding cycles and prey capture in Eudendrium racemosum (Cavolim,~1785). J. Exp. Mar. ±Biol. Coryne (active predators). Ecol., 115:281-293. Most thecates are passive predators. AglaopheniaBoero F. and J. Bouillon. - 1989 The life cycles of Octotiara rus- selliand Stomotoca atra (Cnidaria, Anthomedusae, Pandeida). octodonta is exceptional since its active feedingZoo;. pat- Scr., 18 (I): 1-7. tem is not performed by using mechanoreceptors,Boero F J Bouillon and C. Grayili. - 1991. The life cycle of Hydrichthis minis (Cmdaria: Hydrozoa: Anfhomedusae: Pan- and might be labelled as "random active". The waterdeidae/ Zoo/. J. Linn. Soc., 101:' 189-199. flow caused by beating tentacles suggests that Boero'F-some C.Gravili, F. Denitto, M.P. Miglietta and J. Bouillon. - 1997. The rediscovery of Codonorchis octaedrus species of hydroids, always considered as passive(Hydroidomedusae, Anthomedusae,Pandeidae), with an update filter feeders, can behave like active (but not ciliary)of the Mediterranean hydroidomedusan biodiversity. Ital. J. Zool., 64: 359-365. filter feeders such as barnacles. BoeroF and M. Sara. 1987 Motile sexual stages and evolution Mechanoreceptors are known in some Capitata^ ofLeptomedusae (Cnidaia). Boil. Zool., 54 (2): 131-139: Brindcmann-Voss, A.^- 1970. Anthomedusae/Athecata (Hydrozoa, only and they possibly evolved only once. Unidirec-Cmdaria) of the Mediterranean. Part I. Capitata. Fauna e Flora tional bending as a response to mechanical stimuli,Golfo di Napoli, 39: 1-96. Bumett,_A.L, T.Lentz and M. Wan-en. - 1960. The nematocyst of but irrespective of their direction, on the other hand,Hydro (Part I). The question of control of the nematocyst dis-

70 M.P. MIGLIETTA et al. charge reaction by fully fed Hydro. Annls Soc. r. zoo/.Mackie, Belg., 90:G.O. - 1971. Neurological complexity in medusae: a report 247-267. of central nervous organization in Sarsia. In: Actas I. Simposio Clark, S.D. and C.B. Cook. - 1986. Inhibition of nematocyst int.dis- Zoofilogenia. Acta salmant. (Cienc.), 36:269-280. charge during feeding in the colonial hydroid HalocordyleMarfenin, N.N. dis- - 1981. Nekotorye osobennosti pishchevareniya v ticha (=Pennaria tiarelld): the role of previous prey-killmg.gidrantakh u razlichnykh kolonial'nykh gidroidov. Some obser- Biol. Bull. mar. biol. Lab. Woods Hole, 171: 405-416. vations of the nutrition and the digestive process in hydranths Coma, R., J.M. Gili and M. Zabala. - 1995. Trophic ecology of differenta colonial hydroids. Th. obshch. Biol., 42: 399-408. benthic marine hydroid, Campanularia everta. Mar: Ecol.Mills, C.E. - 1981. Diversity of swimming behaviors in hydrome- Prog.Ser., 119: 211-220. dusae as related to feeding and utilization of space. Mar. Biol., CosteUo, J.H. - 1988. Laboratory culture and feeding of theBerl., 64: 185-189. hydromedusa Cladonema califomicum Hyman_(Anthpmedusa:Orlov, D. - 1997. Observations on the White Sea hydroid, Sarsia C'ladonemidae). J. exp. mar. Biol. Ecol., 123: 177-188. producta (Wright, 1858) (Hydrozoa: Athecata). Sarsia, 81: Costello, J.H. - 1992. Foraging mode and energetics of hydrozoan329-338. medusae. In: J. Bouillon, F. Boero, F. Cicogna, J.M. GiliPassano, and L.M. - 1957. Prey-predator recognidon in the lower Inver- R.G. Hughes (eds): Aspects of Hydrozoan biology, Sci. Mar.,tebrates. In: B. Scheer (ed.): Recent Advances in invertebrates 56: 185-191. physiology, pp. 37-49. Costello, J.H. and M.D. Ford. - in press. Comparison of oblatePiralno, and S., J. Bouillon and F. Boero. - 1992. Halocoryne epizoica prolate medusan propulsion: do oblate medusae use(Cnidaria, jet propul- Hydrozoa), a hydroid that 'bites'. In: J. Bouillon, F. sion? Sci. Mar. Boero, F. Cicogna, J.M. Gili and R.G. Hughes (eds.). Aspects of Edwards, C. and S.M. Harvey. - 1983. Observations on the Hydrozoan Biology, Sci. Mar., 56: 141-147. hydroids Coryne pintneri and Thecocodium brieni newPiraino, to the S., C. Todaro, S. Geraci and F. Boero. - 1994. Ecology of British Ust. /. mar. biol. Ass. U.K., 63: 37-43. the bivalve-inhabiting hydroid Eugymnanthea inquilina in the GiU, J.M., R.G. Hughes and V. Alva. - 1996. A quantitative studycoastal sounds of Tararito (lonian Sea, SE Italy). Mar. Biol., of feeding by the hydroid Tubularia larynx Ellis and Solander,118 (4): 695-703. 1786. Sci. Mar., 60 (I): 43-54. Rees, J.T. - 1979. The polyp and immature stages of Cladonema Grosvenor, W. and G. Kass-Simon. - 1996. Feeding behaviourcalifornicum in Hyman, 1'957 (Anthomedusae: Cladonenudae) Hydro. I. Effects of Artemia homogenate.on nematocyst dis-with biological notes and a discussion of the of the charge. Biol. Bull. mar. biol. Lab. Woods Hole,_Y13^ 527-538.genus Cladonema. J. not. Hist., 13:295-302. Hadrys, H., B. Schierwater and W. Mrowka. - 1990. The Tardent,feeding P. and V. Schmid. - 1972. Ultrastmcture of mechanore- behaviour of a semi-sessile hydromedusa and how it is affectedceptors of the polyp Coryne pintneri (Hydrozoa, Athecata). by the mode ofreproduction.Anim. Behav., 40: 935-944. Exp. Cell Res., 72: 265-275. Hyman, L.H. - 1940.'0bservations and experiments Vannucci,on the physi- M. - 1956. Biological notes and description of new ologyofmedusae. Biol. Bull. mar. biol. Lab. Woods Hole,species 79: of Dipurena (Hydrozoa, Corynidae). Proc. Zoo/. Soc. 282-296. Land., 127: 479-487. Jennings, H.S. -1906. Behavior of the lower organisms.Zamponi, Chapter M.O. XI - 1985. La alimentacion de algunas especies de - Introduction and behavior of Coelenterata. Columbia Univ.Hydromedusae. Neotropica, 31: 155-162. Press, N.Y.: 188-232.

HYDROZOAN ETHOLOGY 71